There is a QR code in the side seam of a Bellver polo. Scan it and you get a page: the fibre, where it was grown, the mill that spun it, the production run this particular garment belongs to, the certificates covering that run and the dates they expire, care instructions for this fabric rather than a row of generic symbols, and a way to get the garment home if somebody finds it on a bus.
Nobody required that. There is no rule in force today obliging a school uniform in Spain to carry any of it. We built it anyway, and I want to be straight about why, because 'we are ahead of the regulation' is the sort of sentence that usually means somebody has read a headline.
What a passport is, underneath the acronym
The EU calls this a Digital Product Passport. Strip away the language and it asks for five things. A permanent identity for the product, reachable through something physical you can scan. Provenance that is structured and machine readable, not a paragraph of prose. Claims that point at evidence, with an honest note where evidence does not exist. Corrections that create a new revision rather than quietly overwriting the old one. And a firm wall between the product data, which is public, and the personal data of whoever owns it, which is not.
That last requirement is why the design took longer than we expected. A code that any stranger can scan, and an owner who must never become visible to that stranger, are two demands pulling in opposite directions. Resolving them properly took an isolated service, an audit trail and a sealed response, which is a lot of engineering for a lost jumper. It is also the only version I would put in a child's clothing.
The timeline, stated properly
The machinery is live. The Ecodesign for Sustainable Products Regulation came into full application on 19 July 2026, and the Commission's central passport registry went operational the day after. The rules for textiles are not live. The working plan gives an indicative 2027 target for adopting the textiles delegated act, and a delegated act cannot apply earlier than eighteen months after entering into force, which puts realistic mandatory compliance for clothing at 2028 or later. The first ESPR delegated acts have already slipped once. I would assume they can slip again.
So no, a Bellver uniform is not complying with anything yet. It is doing early, and voluntarily, what the regulation will eventually describe.
Why we started, which was not compliance
We began this the year my son's skin reacted to a school uniform. I wanted to know what had been against him for six hours a day, and I could not find out. Twenty years in textiles, a training at Polimoda, contacts in half the mills in Europe, and I still could not get a straight answer about a blended polo from a catalogue. That is the question that built the system. The regulation arrived afterwards and asked for the same thing in more formal language.
What it costs to work this way
It is slower. You cannot buy fabric on the open market when a delivery slips, because unrecorded fabric has no story for a garment to inherit. You lose suppliers, not usually because they are hiding something, but because tracing a yarn back through three intermediaries is work nobody was paying them to do before.
And you have to be willing to publish gaps. Some of our records state that fibre origin is not available at that level of detail, because the supplier genuinely cannot tell us and I will not invent a farm to fill a field. A brochure never has gaps. A record does. That is usually how you can tell which one you are reading.
If the textile act is adopted in 2027 and applies in 2028, we will spend that period deciding which level of detail to expose rather than building anything. That is the whole advantage, and it is a quiet one.